Method for selecting remote sensing data and classification algorithms in crop identification and area estimation

Abstract

The invention discloses a method for selecting remote sensing data and classification algorithms in crop identification and area estimation. The method includes the steps of (1), obtaining sensing data of an objective area and performing pretreatment on the obtained remote sensing data; (2), obtaining ground survey data by performing a ground survey on actual distribution of crops in the objective area to obtain a sample crop distribution diagram; determining a training sample and a testing sample of crop classification through the ground survey data; (3), performing dimension expansion on the preprocessed remote sensing data and simulating to generate a multi-resolution image sequence; (4), classifying the crops through a classification algorithm and estimating the planting area of different crops; (5), analyzing the influence of the spatial resolution on the accuracy of crop identification and area estimation and analyzing the effects of planting percentage and aggregation degree on the accuracy of crop planting area estimation; (6), selecting proper remote sensing data and a proper classification method. The method for selecting the remote sensing data and the classification algorithms in crop identification and area estimation provides a theoretical basis and an experimental base for selecting the remote sensing data and the classification algorithms for areas of different crop planting structures.

Description

technical field [0001] The invention relates to a crop remote sensing identification method, in particular to a method for selecting remote sensing data and classification algorithms in crop identification and area estimation. Background technique [0002] With the rapid development of remote sensing technology, it is now able to provide continuous surface sampling from local, regional to global scales, and remote sensing data with a spatial resolution from 0.61 meters to tens of kilometers, to realize the ground survey from multiple spatial scales. remote sensing observation. Scale variability and sensitivity are increasingly playing important roles in analyzes using remote sensing data. In recent years, multi-scale remote sensing data has been widely used in regional and even global land cover mapping, and people have paid more and more attention to the research on the scale effect of classification accuracy of remote sensing data. With the improvement of spatial resolut...

AI Technical Summary

Problems solved by technology

For a specific research area, how much resolution data is needed to meet a certain classification accuracy; how different classifiers respond to the same resolution; how does the accuracy of the same classifier change at different resolutions; The influence of factors such as the number of crops planted and the degree of aggregation on the accuracy results, etc.; the prior art does not give the theoretical basis and experimental basis for the selection of remote sensing data and classification methods for different crop planting structure areas

Images